Patterns of brain activity in choice or instructed go and no-go tasks.

The goal of this study was to investigate the decision making process for choosing what movements to make. We used electroencephalography (EEG) to investigate patterns of the contingent negative variation (CNV) associated with free-choice decisions to move or abstain, comparing them to conditions where actions were commanded. Our primary hypothesis was that choice tasks would differ significantly from each other and exhibit EEG patterns akin to their command-driven counterparts after the decisions were made, at least, in the 50 ms block of time prior to movement. A secondary analysis evaluated post hoc comparisons of time, in 50 ms blocks, to understand the temporal development of the CNV for each condition. We also conducted an exploratory analysis of EEG event-related desynchronization (ERD) to identify patterns of brain activity associated with the decision-making process. This approach was taken due to the exploratory nature of our hypotheses concerning the spatial and temporal characteristics of EEG activity during these free-choice versus commanded tasks. We studied 12 right-handed healthy volunteers (7 women, mean age 53 years, range 39-73 years) with no prior history of neurological or major psychiatric illness. A CNV paradigm encompassing commanded and choice tasks was devised, with a 2500 ms interval between S1 and S2, while recording EEG and electromyography (EMG). S1 provided full information about the upcoming task, which was to be executed at the time of S2. We assessed CNV and explored whole scalp EEG activity, including both voltage as well as power in the alpha and beta frequency ranges. Clear and similar CNVs were observed for command and choice go tasks prior to the movements, contrasting with near-zero CNVs for the command and choice no-go tasks. Separation of CNVs for command go and no-go tasks occurred around 1600 ms post-S1, and choice CNVs separated about 2150 ms post-S1. Exploratory analysis revealed that beta power provided information about decision and preparation processes much earlier. The left dorsolateral prefrontal cortex (DLPFC) exhibited the initial sign of decision approximately 500 ms post-S1 for all tasks, with subsequent preparation for movement or restraint involving distinct activity in various brain regions. The localization of effects in the left DLPFC was determined by visual analysis of the informative electrode sites. The CNVs separate about 2 s after S1, and it appears that this process represents preparation for movement (or no movement). Exploration of the beta activity suggests an earlier decision process which leads eventually to subsequent task preparation and activation. Choice decisions lag slightly behind command decisions, with the CNV apparently reflecting motor implementation rather than the decision-making process. In a simple motor task with an exploratory analysis, both commanded and choice-based decisions are rapidly initiated in the left DLPFC. While the CNV distinguishes between go and no-go conditions, it primarily appears to signify preparation for implementation of the task following the earlier decision. Further controlled studies will be needed to confirm these results.